The effect of enhanced accessibility of acid sites in micro-mesoporous

Jul 19, 2017 - This paper describes a study of the nature and the accessibility of the acid sites in micro-mesoporous mordenite zeolites obtained by d...
0 downloads 10 Views 3MB Size
Research Article pubs.acs.org/acscatalysis

Effect of Enhanced Accessibility of Acid Sites in Micromesoporous Mordenite Zeolites on Hydroisomerization of n‑Hexane Jana Pastvova,†,‡ Dalibor Kaucky,† Jaroslava Moravkova,† Jiri Rathousky,† Stepan Sklenak,† Maryna Vorokhta,§ Libor Brabec,† Radim Pilar,† Ivo Jakubec,∥ Edyta Tabor,† Petr Klein,† and Petr Sazama*,† †

J. Heyrovský Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, Dolejskova 3, 182 23 Prague, Czech Republic ‡ University of Pardubice, Studentská 95, 532 10 Pardubice, Czech Republic § The Institute of Rock Structure and Mechanics, Academy of Sciences of the Czech Republic, V Holesovickach 94/41, 182 09 Prague, Czech Republic ∥ Institute of Inorganic Chemistry, Academy of Sciences of the Czech Republic, Husinec-Rez, 25068 Rez, Czech Republic S Supporting Information *

ABSTRACT: This paper describes a study of the nature and the accessibility of the acid sites in micromesoporous mordenite zeolites obtained by desilication and dealumination and analysis of their activity and selectivity in the hydroisomerization of n-hexane. Alkaline−acid, acid−alkaline−acid, and fluorination−alkaline−acid postsynthesis treatments were employed for the preparation of micromesoporous mordenites. The FTIR spectra of adsorbed d3-acetonitrille, 27Al MAS NMR, HR-TEM, and N2 adsorption were used for quantitative analysis of the Brønsted and Lewis sites, the coordination of Al atoms, and the textural properties. The alkaline treatment causes desilication, preferably occurring along the crystal defects and resulting in the formation of a secondary mesoporous structure characterized by 5−20 nm cavities and the formation of extraframework (AlEx) species and terminal Si−OH groups. The AlEx species formed by hydrolysis of perturbed or dislodged framework Al easily restrict part of the pseudomonodimensional channel structure of mordenite. The subsequent removal of AlEx by mild acid leaching or simultaneous removal of Si and Al atoms by desilication of fluorinated zeolite result in a micromesoporous structure with a large number of unrestricted channel openings and lead to a large increase in the accessibility of OH groups for n-hexane. Thus, the sequential leaching treatments enable the formation of active acid sites in an environment of nonrestricted microporous channels with simultaneous enhancement of accessibility of the active sites and molecular transport. It is shown that the micromesoporous structure with high concentration of Brønsted sites of enhanced accessibility directs the hydroisomerization reaction toward high yields of branched isomers and shortening of the main 12-ring channels and that the larger numbers of channel openings result in an increase in selectivity, limiting nonselective subsequent cracking reactions. KEYWORDS: isomerization, n-hexane, micromesoporous mordenite (MOR) zeolites, hierarchical structure, desilication hexane.1−4 The acid-catalyzed reaction steps determine the overall reaction rate, while the dehydrogenation/hydrogenation reactions are usually in thermodynamic equilibrium and practically do not affect the reaction rate of the hydroisomerization.4,5 Acid sites are located in mordenite zeolites in the small side pockets with eight-membered-ring (8-ring) openings and the main large straight channels with 12-ring openings (Figure 1) forming different local reaction environments affecting the confinement of a reactant molecule and

1. INTRODUCTION The isomerization of linear or low-branched alkane hydrocarbons via an intermediate alkene is an important process in oil refining for production of automotive fuels. The hydroisomerization of n-hexane transforms the linear isomer with low octane number into branched hexanes with sufficiently high octane numbers: e.g., the research octane numbers are 26 and 93 for n-hexane and 2,2-dimethylbutane, respectively. The process is typically catalyzed by bifunctional Pt/H-mordenite catalysts, and its complex mechanism consists of dehydrogenation of hexane to a hexene molecule on metallic platinum, skeletal isomerization of the hexene molecule on an acidic Brønsted site, and rehydrogenation of the branched hexene to © XXXX American Chemical Society

Received: May 24, 2017 Revised: July 17, 2017 Published: July 19, 2017 5781

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

Figure 1. Illustration of the size-restriction of 8-ring in comparison to 12-ring channels in skeletal isomerization of n-hexane over mordenite zeolite: (A) topology of microporous mordenite zeolite consisting 12-ring (7.0 Å × 6.5 Å) channel and 8-ring (5.7 Å × 2.6 Å) side pockets; B) kinetic diameters of hexane isomers;12 (C) reaction scheme.

zeolites can be prepared by a variety of synthesis or postsynthesis methods, e.g., by confined crystal growth,15,23 using polymers as mesoporogens,13 or through postsynthesis desilication or dealumination processes.11,18,24−26 In the present work, we attempted to enhance the accessibility of the active sites by introducing secondary mesoporosity into mordenite zeolites coupled while attempting to preserve a large number of acid sites. The methods for effective formation of secondary mesoporosity through postsynthesis alkaline, alkaline−acid, acid−alkaline−acid, and fluorination−alkaline leaching procedures of conventional mordenite zeolites were explored in the preparation of micromesoporous zeolites with high concentrations of readily accessible Brønsted acid sites. In contrast to dealumination, controlled alkalinebased treatments of the mordenite framework lead to better preservation of the concentration of Brønsted sites and formation of well-developed secondary mesoporosity. The process was followed with respect to the state of Al and the concentration and accessibility of Brønsted sites and Al Lewis sites together with changes in the micromeso texture and how these parameters are reflected in n-hexane hydroisomerization over H/Pt-MOR.

corresponding adsorption enthalpy and stabilization of transition states.6,7 The reaction rate of hydroisomerization of n-hexane over mordenites is thus controlled by the concentration of acid sites located in the 8-ring channel pockets and in the 12-ring main channels with 5 times higher turnover frequency (TOF) per proton in 8-ring channel pockets in comparison with that in 12-ring channels.8 Nevertheless, the accessibility of 8-ring (2.6 Å × 5.7 Å) in comparison to 12-ring (6.5 Å × 7.0 Å) channels is diffusionrestricted for n-hexane (cf. the size of the channel openings and kinetic diameters of the hexane isomers in Figure 19−12) and the efficiency of the catalytic process is limited by mass transfer effects.5,13 Because of their larger kinetic diameter, the branched hexanes diffuse even more slowly than n-hexane.14 The accessibility of acid sites in the current industrial catalysts is enhanced by acid-treated partially dealuminated mordenites with improved textural properties.5 However, dealumination leads to a decrease in the concentration of Brønsted sites, the formation of Lewis sites, and small changes in the micropores.5 In recent years, considerable efforts have been devoted to improving the pore accessibility and molecular transport of reactants to active sites in micropores, namely in MFI zeolites, to enhance the activity in yielding the desired products in a variety of important acid-catalyzed reactions.15−20 Several approaches have been developed to shorten the length of micropores and thus enhance mass transport by using zeolite nanosheets and nanocrystals17,21,22 or zeolites that contain both micro- and mesopores. Micromesoporous (hierarchical)

2. EXPERIMENTAL SECTION 2.1. Preparation of Micromesoporous H- and Pt/HMordenite Zeolites. Mordenite zeolite (Zeolyst Int., CBV 20A molar Si/Al 12.1), denoted as MOR/12, was used for the 5782

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

Figure 2. Scheme of the preparation of micromesoporous mordenite zeolites using postsynthesis alkaline−acid (series I), acid−alkaline−acid (series II), and fluorination−alkaline−acid (series III) leaching procedures.

zeolites by incipient wetness impregnation with a H2PtCl6 solution to yield 1.5 wt % of Pt. The Pt-impregnated granulated zeolite was activated before the catalytic test in a stream of O2 at 450 °C for 3 h, purged by a N2 stream at 450 °C, followed by cooling to 250 °C, and finally activated in a mixture of 80 mol % of H2 and 20 mol % of N2 at 250 °C for 1 h. 2.2. Characterization of Zeolites. A Bruker AXSD8 Advance diffractometer with Cu Kα radiation in Bragg− Brentano geometry with a graphite monochromator and a position-sensitive detector (Våntec-1) was used to record the X-ray powder diffraction (XRD) patterns. The porosity of the modified mordenites was determined by analysis of the adsorption isotherms of nitrogen at the boiling point of liquid nitrogen (77 K). Before the adsorption experiment, the samples were outgassed at 240 °C for at least 24 h to ensure complete cleaning of the surface. The experiments were carried out using an ASAP 2010 apparatus (Micromeritics). The volume of micropores and the external surface area were determined by the Broekhoff−de Boer t plot. The total pore volume was determined by converting the adsorption at the relative pressure of 0.8 to the volume of liquid nitrogen. The relative pressure around 0.8 for practically all of the investigated samples is high enough for complete filling of the pores and low enough to have substantial proportion in the interstitial space. The volume of mesopores was finally calculated by subtracting the volume of micropores from the total pore volume (at the relative pressure of 0.8). The crystal morphology was analyzed using a JEOL JSM-5500LV scanning electron microscope (SEM), (Figure S1 in the Supporting Information). Highresolution transmission electron microscopy (HR-TEM) was carried out on a JEOL JEM 3010 microscope operated at 300 kV (LaB6 cathode, point resolution 1.7 Å). The length of the diffusion path in the mordenite channel in the crystal with the mesopores was estimated using a model approach as follows: the crystal size is 80 × 80 × 80 nm, and the volume

preparation of micromesoporous mordenites using postsynthesis alkaline, alkaline−acid, acid−alkaline−acid, and fluorination−alkaline−acid leaching procedures (Figure 2) and as a standard for comparing catalytic properties. The first series of micromesoporous mordenites was prepared using alkaline and alkaline−acid treatments by leaching MOR/12 in an alkaline solution (30 mL of 0.2 M NaOH per 1 g of mordenite stirred in a beaker at 85 °C for 2 h, sample mm-MOR/8)27 and subsequent leaching in acid solutions (30 mL of 0.1 M oxalic acid per 1 g of alkaline-treated zeolite stirred in a beaker at 85 °C for 20 h, sample deAl-mm-MOR/9, or 40 mL of 0.1 M HNO3 per 1 g of alkaline-treated zeolite stirred in a beaker at 50 °C for 15 min, sample deAl-mm-MOR/9.5). The second series was prepared using acid−alkaline−acid treatments with dealumination of MOR/12 (30 mL of 3 M HNO3 per 1 g of mordenite stirred and heated under reflux at 80 °C for 40 min, sample deAl-MOR/16), subsequent desilication (30 mL of 0.2 M NaOH per 1 g of dealuminated zeolite stirred in a beaker at 85 °C for 2 h, sample mm-deAl-MOR/9.1), and finally mild dealumination (30 mL of 0.1 M HNO3 per 1 g of zeolite at 50 °C for 15 min, sample deAl-mm-deAl-MOR/13, or 30 mL of 0.1 M (COOH)2 per 1 g of zeolite at 85 °C for 20 h, sample deAl-mm-deAl-MOR/30). The third series was prepared by fluorination of MOR/12 (5 g of mordenite was impregnated by 15 mL 0.36 M NH4F at room temperature, stored at room temperature for 8 h, and subsequently dried at 120 °C for 12 h), subsequent desilication (30 mL of 0.2 M NaOH per 1 g of fluorinated mordenite stirred in a beaker at 85 °C for 30 min) and calcination at 550 °C for 3 h (sample mm-F-MOR/9.4), with subsequent leaching in acid solutions (30 mL of 0.1 M HNO3 per 1 g of zeolite at 50 °C for 15 min, sample deAl-mmF-MOR-F/13.6). All of the zeolites were ion-exchanged with 0.5 mol dm−3 NH4NO3 at room temperature (1 g of a zeolite per 100 cm3 of solution, three times over 12 h). Pt was introduced into the 5783

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

Figure 3. Representative HR-TEM images of (A, D) microporous MOR (MOR/12), (B) micromesoporous MOR (mm-MOR/8), (C) dealuminated micromesoporous (deAl-mm-MOR/9.5), (E) fluorinated MOR (F-MOR), and (F) fluorinated micromesoporous MOR (mm-FMOR/9.4).

corresponding to the mesopores is 0.09 cm3 g−1. Two cases were modeled: either 18 mesopores of 8 nm diameter or 72 mesopores of 4 nm diameter. The mesopores of length 80 nm were randomly positioned and oriented along one of the crystal axes. Three cases of diffusion paths were considered: the mordenite channel connects (i) one mesopore with the external surface of the crystal, (ii) two mesopores, and (iii) two external surfaces. One million simulations were used employing a pseudorandom number generator. The chemical compositions of the parent and prepared Pt-zeolites were determined by Xray fluorescence spectroscopy using a PW 1404 instrument (Philips). The FTIR spectra of the zeolites evacuated at 450 °C for 3 h were recorded at room temperature on a Nicolet Nexus 670 FTIR spectrometer operating at 2 cm−1 resolution by collecting 256 scans for a single spectrum. The adsorption of d3-acetonitrile (13 mbar of CD3CN at room temperature for 20 min with subsequent evacuation for 15 min at room temperature) was used for analysis of the H-mordenites. The characteristic IR bands of the CN vibrations and the extinction coefficients εB = 2.05 cm μmol−1 and εL = 3.60 cm μmol−1 28 were used for determination of the concentrations of acidic Brønsted and Lewis sites. The adsorption of n-hexane (1.4 Torr of n-hexane at room temperature for 20 min with gradual desorption to 0.005 Torr at room temperature) was used to estimate the accessibility of the OH groups of the mordenites for n-hexane. The difference between the intensity of the absorption bands of the OH groups in the spectra of the dehydrated mordenites and the spectra after adsorption of nhexane provided information on the accessibility of the OH groups in mordenite. The desorption of n-hexane from the samples of microporous (MOR/12) and micromesoporous (deAl-mm-MOR/9.5) mordenites was followed by using a Nicolet 6700 FTIR spectrometer and an IR cell equipped with NaCl windows and connected to a vacuum line and n-hexane-

dosing setup. The self-supported wafers with thickness 90%), (Table 2). The absence of mesopores, the increased accessibility of the OH groups, and the presence of internal silanol groups indicate that extraction of the framework Al atoms primarily induced changes in the distribution of the sizes and/or the connectivity of the microporous channels. However, in addition to appreciable textural changes, these structural changes caused a significant decrease in the concentration of the Brønsted acid sites to about half and an increase in the ratio of Lewis/Brønsted sites. 3.1.2. Effect of Desilication on Structure of Mordenite. TEM images of the alkaline-leached zeolite mm-MOR/8 (Figure 3) clearly show that the treatment resulted in extensive formation of a secondary mesoporous structure characterized by numerous cavities about 5−20 nm in size in the microporous crystals. The adsorption isotherms of desilicated mm-MOR/8 zeolite (Figure 4) exhibit a hysteresis loop typical for mesopores and adsorption in the zeolite micropores with a significant decrease in the volume of the micropores from 0.19 to 0.06 cm3 g−1 (Table 1). The preservation of the crystalline structure of mm-MOR/8, which is obvious from the X-ray diffraction patterns (Figure S2 in the Supporting Information) 5787

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

Figure 6. Schematic illustration of the effect of the extraction of one T atom from an 8-ring on the local structure of the openings of the side pockets in MOR zeolite: (A) two neighboring 8-rings and the same structure with removed (B) T4 atom and (C) T2 atom.

Mordenite. Our recent study19 on the formation of secondary mesoporosity through postsynthesis sequential leaching procedures demonstrated that the Brønsted active sites located in the environment of nonrestricted micropores and accessible through mesopores can be obtained by alkaline and subsequent mild-acid leaching. Treatment of mm-MOR/8 using 0.1 M oxalic acid under mild conditions (deAl-mm-MOR/9) resulted in a very slight increase in the molar Si/Al ratio and a significant increase in the micropore volume from 0.06 to 0.13 cm3 g−1 that further extended the volume of mesopores (Table 1), a significant increase in the concentration of Brønsted sites from 0.44 to 0.81 mmol g−1 and Levis sites from 0.14 to 0.35 mmol g−1 (Table 2), disappearance of the lowering symmetry of the signal at 55 ppm in the 27Al MAS NMR spectra (Figure S4 in the Supporting Information), and elimination of the absorption band at 3650−3670 cm−1 due to Al−OH groups (Figure 5). These findings clearly indicate that the extraframework and/or perturbed framework Al species formed in mesopores after desilication and blocking part of the micropores were removed by acid leaching. The increase in the concentrations of the acid sites indicates that the leaching selectively removed the extraframework tetrahedrally coordinated AlOx/(OH)y species blocking the micropores, whereas the framework Al atoms associated with bridging OH groups were barely affected. Acid leaching of mm-MOR/8 using 0.1 M nitric acid resulted in slightly more pronounced structural changes in comparison to the treatment using oxalic acid. Nitric acid, as a stronger dealumination agent,51 caused a moderately greater increase in the molar Si/Al ratio, the same increase in the micropore volume from 0.06 to 0.13 cm3 g−1, and further extension of the mesoporous volume to 0.12 cm3 g−1 (Table 1). A lower concentration of Brønsted sites (0.55 mmol g−1) indicates that part of the framework Al atoms was extracted from the framework and/or perturbed by the treatment in nitric acid. The removal of extraframework Al in the deAl-mm-MOR/9 sample by oxalic acid caused only minor changes in the availability of OH groups in comparison to the parent zeolite (Table 2). A significant increase in the accessibility of the OH groups (∼80%) occurred after slight dealumination of the zeolite framework in the deAl-mm-MOR/9.5 sample by nitric acid. It seems that desilication with subsequent removal of extraframework Al can lead to shortening of the main 12-ring channels, but the number of new channel openings created in

mesopores is not large enough for significant enhancement of the accessibility of the OH groups in 8-ring side pockets. Partial dealumination of mm-deAl-MOR/9.1 using 0.1 M nitric acid caused similar changes to the dealumination of mmMOR/8. However, they are less pronounced: i.e., opening of restricted micropores with an increase in the volume of micropores from 0.15 to 0.19 cm3 g1 and increase in the concentration of Lewis sites from 0.22 to 0.45 mmol g−1. The concentration (0.43 mmol g−1) and accessibility of Brønsted acid sites (∼92%) are very comparable with those of deAlMOR/16. Thus, the desilication of deAl-MOR/16 with subsequent removal of AlEx by mild acid leaching resulted in deAl-mm-deAl-MOR/13 with well-developed mesoporous structure and with similar concentration and accessibility of OH groups for n-hexane as the starting deAl-MOR/16 before the subsequent alkaline−acid treatments. The hydrolysis of Al−O and Si−O bonds in the 8-ring (formed by T2 and T4 atoms) during acid and alkaline treatments results in dislodgement of the T atom from the framework position and dramatic changes in the local arrangement. Figure 6 illustrates how the openings of two neighboring side pockets in MOR zeolite can be altered by the removal of one T atom in the T2 or T4 sites. The extraction of the T4 atom leads to a merging of the two 8-rings, providing a large void space enabling a diffusion of large molecules: e.g., benzene.52 The extraction of the T2 atom could possibly provide a large deformed elongated ring; however, previous reports indicate that the extraction is site specific and the dealumination of T4 is strongly favored.52−54 The creation of these open connections by the preferable extraction of Al atoms from the T4 sites explains the dramatic increase in the accessibility of the bridging of OH for hexane even after very mild dealumination of the mordenite framework. 3.1.4. Effect of Fluorination−Alkaline Treatment on the Porous Structure and Accessibility of Acid Sites in Mordenite. Recently, Yu et al.55 showed that impregnation of microporous ZSM-5 zeolite with an aqueous solution of NH4F prior leaching by a solution of NaOH increases the susceptibility of the zeolite toward leaching of framework Al atoms and provides controlled extraction of both Al and Si atoms. We modified the procedure for alkaline treatment of fluorinated mordenites and obtained the mordenite zeolite mm-F-MOR/9.4 with the mesopore volume increased to 0.09 cm3 g−1 and the volume of the micropores concurrently slightly decreased to 0.14 cm3 g−1, 5788

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

Figure 7. Analysis of the effect of the presence of mesopores on the length of zeolite channels and diffusion of n-hexane: (A) model determining the length of the diffusion path through the mordenite channel, where 4 and 8 nm channels oriented along the axes and internal mesopore volume 0.09 cm3 g−1 were considered; (B) desorption of n-hexane initiated by changes in the n-hexane partial pressure from 2.5 to 1.5 Torr analyzed by FTIR spectroscopy.

controlled Al and Si extraction during the alkaline treatment. The finding that this treatment significantly improved the accessibility of OH groups for n-hexane is of great importance. The simultaneous removal of Si and Al atoms and smaller size of part of the formed mesopores with possibly a larger number of new channel openings results in a large increase in the accessibility of the OH groups for n-hexane. 3.1.5. Effect of Mesoporosity on the Pore Length. In order to illustrate the effect of mesoporosity on the reactant transport, a model showing the shortening of the main 12ring channels was made and the desorption of n-hexane was analyzed for the microporous and a representative micromesoporous mordenite (Figure 7). The model is based on the obtained SEM, HR-TEM, and N2 adsorption data considering the mesoporous volume 0.09 cm3 g−1, which is approximately 15% of the total volume of the crystal, and the crystal size of 80 nm. Two cases were modeled with randomly positioned mesopores of diameter 4 or 8 nm oriented along one of the crystal axes and connected to the external surface of the crystal. The mesopores that communicate with the external surface of the mordenite crystal are observed in the high-resolution TEM images, and the spatial arrangement of the mesopores in the entire void volume of the crystal was demonstrated by Milina et al.58 using positron annihilation lifetime spectroscopy for micromesoporous zeolites prepared by desilication using the same amount and concentration of NaOH solution. Three cases of the diffusion paths were considered: the mordenite channel connects (i) one mesopore with the external surface of the crystal, (ii) two mesopores, and (iii) two external surfaces (i.e., the diffusion path 80 nm) if there is no mesopore present. One million simulations using a pseudorandom number generator yielded diffusion paths of 42 and 39 nm for the mesopores of the 8 and 4 nm diameters, respectively. The diffusion path depends only slightly on the diameter of the mesopores, since only approximately 15% of the crystal is

indicating the formation of secondary mesoporosity and opening of nonrestricted channel openings of the preserved microporous structure. In addition to the small, near-circular bright areas reflecting formation of intracrystalline mesoporosity, HR-TEM images (Figure 3) also show mesopores formed along boundaries and intergrowths. The dissolution or etching along boundaries and intergrowths and the indiscriminate dissolution of Al and Si are characteristic of chemical etching with fluoride ions in NH4F-HF,56 and Al-directed dissolution of siliceous areas is typical for alkaline treatment.57 Two modes of the formation of the mesoporous structure were also reported by Svelle et al.50 for standard desilication. They demonstrated the creation of mesopores as a consequence of both Al-directed dissolution of siliceous areas and selective dissolution or etching along boundaries, intergrowths, and defects within each particle. HR-TEM images of mm-F-MOR/9.4 show concomitant contribution of both modes of mesopore formation. The slight shift of the maximum of the peak at about 55 ppm in the 27 Al MAS NMR spectra (Figure S4 in the Supporting Information) reflects changes in the coordination and local surroundings of the Al sites and/or a site-specific dealumination of the individual T sites.52−54 The concentration of Brønsted and Lewis sites, 0.6 and 0.24 mmol g−1, respectively, determined by adsorption of d3-acetonitrile, and high accessibility (>90%) of bridging OH groups for n-hexane molecule (Table 2) for the mm-F-MOR/9.4 sample show that the desilication of fluorinated mordenite zeolite yields a micromesoporous structure with high concentration of readily accessible acid sites. The development of secondary mesoporosity and preservation of microporosity via concurrent extraction of SiO2 and a smaller amount of Al is enabled by transformation of a fraction of Si−O−Al into the framework as F-bearing tetrahedral Al entities after fluorination.55 The presence of F-bearing tetrahedral Al species which are readily dislodged in alkaline medium55 and Al sites in Si−O−Al entities with persistent resistance to alkaline treatment enabled 5789

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

Figure 8. Yields of branched C6 isomers and C1−C5 byproducts in the hydroisomerization of n-hexane over Pt/H-MOR zeolites as a function of temperature: (A, B) series I; (C, D) series II; (E, F) series III.

practically independent of the surface of the metallic platinum clusters for platinum loadings from 0.5 to 6 wt % in Pt-faujasite zeolites. Consistently, Van de Runstraat et al.61 showed that the hydrogenation/dehydrogenation functions of platinum was not rate-determining for Pt/H-*BEA with a platinum loading of ∼1.5 wt %. Accordingly, to ensure that the dehydrogenation/ hydrogenation reactions do not limit the overall alkane hydroisomerization, all of the Pt/H-MOR catalysts were prepared with a concentration of 1.5 wt % Pt and the dispersity of the platinum clusters was checked using HR-TEM and FTIR spectra of adsorbed CO. Well-dispersed platinum clusters with particle sizes ∼1−20 nm located on the external surface of the zeolite crystals were observed in the HR-TEM images of all the microporous and micromesoporous Pt-MOR zeolites (Figure S7 in the Supporting Information). The broad distribution of the Pt cluster sizes is typical for Pt-zeolites prepared by conventional incipient wetness impregnation. Similar intensities of the absorption bands of the hydroxyl group in the FTIR spectra before and after impregnation of the mordenites with platinum indicate that the Pt reduced under a hydrogen atmosphere is not coordinated in cationic sites to a significant extent (not shown). The FTIR spectra of CO adsorbed on microporous and micromesoporous Pt/H-MOR zeolites (Figure S8 in the Supporting Information) exhibit the presence

removed and a majority of the crystal is not influenced by the mesopores. The series of FTIR spectra recorded during transient desorption of n-hexane induced by a very fast decrease in the partial pressure of n-hexane from 2.5 to 1.5 Torr are shown for MOR/12 and deAl-mm-MOR/9.5 in Figure 7B. The partial pressure of 1.5 Torr provides the loading of one n-hexane molecule per acid site, while n-hexane molecules weakly interacting with the inert channel walls are also present at the partial pressure of 2.5 Torr. Under such a decrease in the partial pressure, the changes in transport can be associated with the alterations of the pore structure rather than with desorption of n-hexane from the acid sites. The higher desorption rate observed for the micromesoporous deAl-mm-MOR/9.5 in comparison to microporous MOR/12 is thus consistent with the effective reduction of the crystal domain by the introduction of the mesopores communicating with both the micropores and connected with the external surface. The enhanced transport in the micromesoporous zeolite can also be, however, affected by a contribution from a complex interplay of changes in the local structure of microporous channels and the channel openings.59 3.1.6. Structure of Pt Species. Ribeiro et al.60 showed that the reaction rate in the hydroisomerization reaction is 5790

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

Table 3. Yields of Branched Hexane Isomers and Lower Molecular Weight Byproducts for Hydroisomerization of n-Hexane over Microporous and Micromesoporous Pt/H-MOR at 225 and 250 °C ∑yiso‑C6 (%)

2,2-dimethyl (%)

2,3-dimethyl (%)

2-methyl (%)

3-methyl (%)

C1−C5 (%)

sample

225 °C

250 °C

225 °C

250 °C

225 °C

250 °C

225 °C

250 °C

225 °C

250 °C

225 °C

250 °C

Pt/H-MOR/12 Pt/H-mm-MOR/8 Pt/H-deAl-mm-MOR/9 Pt/H-deAl-mm-MOR/9.5 Pt/H-deAl-MOR/16 Pt/H-mm-deAl-MOR/9.1 Pt/H-deAl-mm-deAl-MOR/13 Pt/H-deAl-mm-deAl-MOR/30 Pt/H-mm-F-MOR/9.4

36.5 40.1 37.5 48.7 54.1 45.8 50.0 23.7 62.5

53.0 59.2 61.9 67.7 60.8 63.4 67.5 47.9 59.6

1.6 2.5 2.1 1.8 2.9 2.8 2.4 0.6 4.9

3.6 5.2 5.4 5.7 4.9 8.1 6.3 3.5 7.4

6.0 5.2 4.7 6.1 6.1 4.5 5.0 1.3 7.4

6.0 6.3 6.7 7.3 6.8 6.5 7.4 4.2 6.2

7.3 7.9 8.7 8.8 11.2 8.6 8.2 7.3 8.6

8.4 8.6 8.7 10.5 11.3 8.6 10.0 8.4 8.4

21.6 24.5 22.0 32.0 33.9 29.9 34.4 14.5 41.6

35 39.1 41.1 44.2 37.8 40.2 43.8 31.8 37.6

0.5 0.9 0.6 3.0 3.9 2.9 3.7 5.0 6.4

5.3 7.5 4.1 7.6 16 10.4 8.7 12.4 18.5

of a broad absorption band at 2078 cm−1 characteristic for reduced Pt metallic clusters 1−20 nm in size62,63 located on the external surface of the zeolite crystal. The reduction of Pt into the zerovalent oxidation state of metallic Pt is also indicated by the absence of the absorption bands at 2170−2150 cm−1 and at 2130−2140 cm−1 characteristic of CO interacting with Pt2+ counterions and Ptnδ+ clusters, respectively.62,63 Thus, all of the Pt/H-MOR contains well-dispersed platinum in the form of metallic clusters that can provide sufficient hydrogenation/ dehydrogenation reactions occurring for the employed Pt loadings much more rapidly than isomerization reactions on acid sites, regardless of the specifics of their structure for all of the Pt/H-zeolite catalysts.60,64 The presence of the secondary mesoporous structure can possibly increase the proximity between the metallic platinum and the Brønsted sites. The effect of nanoscale proximity of metallic sites and acid sites on the hydroconversion of hydrocarbons was recently assessed by Zecevic et al.65 They showed that the closest proximity between the metal and the Brønsted acid sites in zeolites does not provide improved catalytic properties in hydrocracking reactions. This is in line with the results of our study on hydroisomerization,66 where we analyzed whether the activity of Pt/H-MOR zeolite is affected by the proximity of Pt species and acid sites using the samples of H-MOR/12-Pt/Al2O3 prepared as mechanical mixtures of the protonic form of the zeolite and Pt/Al2O3. The yields were practically identical for both the Pt/H-MOR and H-MOR/12Pt/Al2O3 with platinum metal deposited on either the zeolite or alumina.66 3.2. Hydroisomerization of n-Hexane. 3.2.1. Hydroisomerization over Microporous, Desilicated, and Partially Dealuminated-Desilicated Mordenites. Figure 8 depicts the yields of branched hexane isomers (2-methylpentane, 3methylpentane, 2,2-dimethylbutane, and 2,3-dimethylbutane) and C1−C5 byproducts (methane, ethane, propane, butanes, and pentanes) as a function of the reaction temperature in the hydroisomerization of n-hexane over the microporous and micromesoporous Pt/H-MOR zeolites. The yields of the individual products at 225 and 250 °C are given in Table 3. With microporous MOR/12, the conversion first started to increase at a temperature of approximately 150 °C and reached a maximum of 58% at approximately 275 °C before declining again due to nonselective cracking reactions, as is indicated by the steadily increasing conversion of n-hexane and yield of C1− C5 byproducts. The desilication of the parent sample resulted in little improvement in conversions over the entire temperature range, passing through a maximum at 250 °C with the yields increasing from 53.0 to 59.2% and from 9.6 to 11.5% for all of

the branched and dibranched hexanes, respectively, and then declining due to nonselective cracking at higher temperatures. The catalytic behavior of the mm-MOR/8 sample is in accordance with the findings of Monteiro et al.,67 showing that the desilication of mordenite zeolite does not significantly improve the catalytic efficiency of the mordenite zeolites. With the deAl-mm-MOR/9 sample obtained by the removal of extraframework Al from mm-MOR/8 using oxalic acid, the formation of byproducts at high temperatures was significantly reduced in comparison to mm-MOR/8 or microporous MOR/ 12 samples and was accompanied by a corresponding increase in the yield of the desired branched isomers, reaching a maximum of 67.4% at 275 °C. The deAl-mm-MOR/9 sample is characterized by shortened 12-ring channels due to a welldeveloped secondary mesoporous structure, but without significantly enhanced accessibility of the OH groups in comparison to the parent zeolite (Table 2). It seems that the shortening of the main channel and a larger number of channel openings provide only a slight increase in activity but lead to a significant increase in selectivity, limiting the nonselective subsequent cracking reactions and consequently increasing the yield of desired products at higher temperatures. The deAl-mm-MOR/9.5 sample dealuminated by nitric acid provides higher yields of branched isomers in comparison to MOR/12 and mm-MOR/8 zeolites with similar yields of byproducts (Figure 8). As the difference between deAl-mmMOR/9.5 and deAl-mm-MOR/9 lies mainly in the significantly higher accessibility of the OH groups (Table 2), the improvements in the yields of branched isomers at lower temperatures, e.g. from 53% to 67.7% at 250 °C, can be associated with greater accessibility of the acid sites and the improvements in the yield at higher temperatures, e.g. from 58.7 to 67.4% at 275 °C, also with the shortened main channels and larger number of channel openings. Thus, the use of the deAl-mm-MOR/9.5 sample had a positive effect on the yield of branched isomers in the whole temperature region. The improvements in the catalytic properties are minor, however, clearly indicating the potential for improvements in the functionality of hydroisomerization mordenite catalysts via enhanced accessibility of active sites and transport of reactants in the micromesoporous structure. The durability and stability of the catalytic activity for the deAl-mm-MOR/9.5 as a representative micromesoporous Pt/ H-MOR zeolite were analyzed under the reaction conditions. Stable values of the yields of mono- and dibranched isomers, ∼55% and ∼9%, respectively, and byproducts ∼1% as a function of time on stream were obtained over deAl-mmMOR/9.5 at 235 °C for 72 h, indicating a sufficient structural 5791

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis

MOR/9.4 sample, e.g. from 13.5% to 47.2% at 200 °C, showing that the desilication of fluorinated mordenite zeolite markedly increased the catalytic activity (Figure 8E). This was accompanied by higher yields of dibranched isomers (Table 3) and high selectivity without the detectable formation of C1− C4 byproducts at ≤200 °C (Figure 8F). In comparison with other micromesoporous Pt/H-MOR, the mm-F-MOR/9.4 sample provides the greatest enhancement in the yields of branched C6 isomers achieved at low temperatures (Figure 10).

stability and no deactivation of the micromesoporous Pt/HMOR (Figure 9).

Figure 9. Dependence of the yield of branched hexane isomers and C1−C3 byproducts on time on stream (TOS) in hydroisomerization of n-hexane over Pt/H-deAl-mm-MOR/9.5 at 235 °C.

3.2.2. Hydroisomerization over Partially Dealuminated and Acid−Alkaline−Acid-Treated Mordenites. Figure 8C shows the yields of branched isomers as a function of temperature obtained over dealuminated deAl-MOR/16 and zeolites prepared by subsequent desilication and desilication/ dealumination of the deAl-MOR/16 sample. Although the dealumination of mordenite (deAl-MOR/16) leads to a decrease in the concentration of Brønsted sites from 1.06 to 0.47 mmol g−1, the accessibility of the remaining acid sites was significantly enhanced, with ∼92% of the Brønsted sites accessible for n-hexane (Table 2). Less than half the concentration of Brønsted sites in deAl-MOR/16 provided yields of branched hexanes which outperform the microporous MOR/12: see, for example, the yields of branched hexanes 36.5% and 54.1% at 225 °C for MOR/12 and deAl-MOR/16, respectively. The higher specific activity of acid-treated partially dealuminated mordenites has been well documented in the literature68 and explained by the beneficial effects on the textural properties of extraction of part of the framework Al atoms by acid treatment, leading to improved accessibility of highly active acid sites located in the 8-ring side pockets.5,69 Desilication of the deAl-MOR/16 sample yielded micromesoporous mm-deAl-MOR/9, which showed no significant changes in the yields of the branched isomers nor byproducts. The positive effect of the development of the mesoporous volume (0.11 cm3 g−1) was probably offset by the negative effect of the decline in the microporous volume because of its partial blockage. The subsequent partial dealumination of mmdeAl-MOR/9 provided the deAl-mm-deAl-MOR/13 sample with a well-developed mesoporous structure and with similar concentration and accessibility of the OH groups for n-hexane in comparison to the deAl-MOR/16 sample. The similar concentrations and accessibility of the OH groups for n-hexane in the deAl-MOR/16 and deAl-mm-deAl-MOR/13 samples resulted in similar yields of branched hexane isomers at ≤225 °C. The presence of mesoporous structures in deAl-mm-deAlMOR/13 caused a slightly lower formation of cracking products and a corresponding increase in the yield of branched hexanes at temperatures ≥250 °C. 3.2.3. Hydroisomerization over Fluorinated-Alkaline Treated Mordenites. The yields of branched C6 isomers in the hydroisomerization with microporous MOR/12 without postsynthesis treatment were very low at temperatures ≤200 °C. The yields were increased considerably over the mm-F-

Figure 10. Comparison of the effects of the postsynthesis alkaline− acid (Pt/H-deAl-mm-MOR/9.5), acid−alkaline−acid (Pt/H-deAlmm-deAl-MOR/13), and fluorination−alkaline (Pt/H-mm-F-MOR/ 9.4) treatments on the yields of branched hexanes at 175 and 200 °C.

The chemical analysis showed the absence of fluorine in the mm-F-MOR/9.4 treated after fluorination by desilication, triple ion exchange with a large excess of NH4NO3 solution, and calcination in air and subsequently under a hydrogen atmosphere. The enhancement in the activity is thus associated with the micromesoporous structure providing a high concentration of Brønsted sites (cB = 0.6 mmol g−1) readily accessible for n-hexane molecules (>90%) (Table 2). To exclude the possible effect of traces of fluoride not detectable by the chemical analysis on the increase in activity, a fluorinated sample of Pt/H-F-ZSM-5 zeolite with small crystals was prepared and its activity was compared with that of Pt/HZSM-5. The intracrystalline diffusion does not affect the overall reaction rates for ZSM-5 for either the n-hexane hydroisomerization61 or cracking.70 Accordingly, we observed very comparable yields of branched C6 isomers over Pt/H-ZSM-5 and its fluorinated analogue Pt/H-F-ZSM-5 in the entire temperature region (Figure S9 in the Supporting Information). Thus, the high concentration of the Brønsted sites located in nonrestricted micropores highly accessible via new channel openings in small mesopores formed by the simultaneous removal of Si and Al atoms results in significantly improved catalytic properties, enabling the reaction at lower temperatures.

4. DISCUSSION 4.1. Structure of Micromesoporous Mordenites Obtained by Leaching Methods. Selective dissolution of silicon-rich areas of the framework and the extraction of aluminum are inherent parts of the acid and alkali treatments of zeolites, respectively. Our results clearly show that, for the formation of micromesoporous mordenite with Brønsted active sites readily accessible in nonrestricted micropores, it is necessary to extract both silicon and aluminum atoms in a controlled way to ensure the formation of mesopores and preservation of micropores and the majority of the framework Al. Particularly for desilication of mordenites with a high 5792

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis concentration of aluminum in the framework (Si/Al ≈ 12), the extraframework tetrahedrally coordinated AlOx/(OH)y species, formed by hydrolysis and dislodgment of a small fraction of framework Al, block a significant part of the pseudomonodimensional 12-ring channels and/or accessibility of the acid sites in 8-ring side pockets. Desilication of mordenite surprisingly does not lead to greater accessibility of the acid sites; to the contrary, a significant part of the micropores is diffusion-limited even for nitrogen molecules. However, subsequent acid leaching under specific conditions, characterized by a low concentration of oxalic or nitric acid, leads to efficient removal of the extraframework and/or perturbed framework Al species formed in mesopores after desilication and blocking of part of the micropores. The concentration of the Brønsted sites determined by d3-acetonitrile and their accessibility for nhexane as a relevant molecular probe for the hydroisomerization reaction indicate that the prevailing amount of the framework Al atoms is not dislodged during either the dissolution of silicon-rich areas of the framework or the selective extraction of the extraframework Al. Thus, the alkaline removal of parts of the Si-rich framework and subsequent selective acid leaching of extraframework Al and only a small amount of framework Al resulted in a micromesoporous mordenite structure with a high concentration of strongly acidic Brønsted sites readily accessible in both 12-ring and 8-ring channels. The desilication with subsequent removal of only extraframework Al creates a number of new channel openings in mesopores, effectively shortening the main 12-ring channels. However, the dealumination of a small fraction of the framework Al atoms is indispensable for significant enhancement of the accessibility of the OH groups in 8-ring side pockets. It should be also mentioned that the increase in the concentration of Lewis centers and an intense signal of internal Si−OH groups suggest that the massive changes in the texture of the zeolite toward a well-developed mesoporous structure leads to the disruption of part of the regularly arranged atoms originally present in the low-defective zeolitic structure. The alkaline leaching of mordenite preimpregnated with a solution of NH4F appeared to be the most efficient approach for creating the secondary mesoporous structure in mordenite zeolite with a high concentration of readily accessible Brønsted sites. This procedure enabled controlled extraction of Si atoms and concomitantly the framework F-bearing tetrahedral Al species. The dissolution of tetrahedrally coordinated Al(FO3) and SiO4 entities during alkaline treatment resulted in secondary mesoporosity and opened nonrestricted channel openings of the preserved microporous structure with a high concentration of Brønsted sites readily accessible for n-hexane. A difference was observed in the modes of formation and types of mesoporous structures obtained by desilication−acid treatment and desilication of fluorinated mordenite. With the desilication−acid treatment, circular mesopores 5−20 nm in size were predominant, while desilication of fluorinated mordenite yielded small mesopores combined with mesopores formed along crystal boundaries and intergrowths. 4.2. Effect of Micromesoporous Structure on Hydroisomerization of n-Hexane. The molecules involved in hydroisomerization as either reactants or products have sizes similar to those of the mordenite pores (Figure 1).14 Consequently, the access of these molecules to active sites is a key factor governing the performance of mordenite catalysts. The most obvious effects of the introduction of secondary mesoporosity on the yields of branched hexanes obtained in

this study is illustrated in Figure 10. Whereas only a small increase in the yield of branched isomers from 13.5% to 16% was obtained after dealumination, the increase in the yield to 22% was appreciable with concurrent introduction of a mesoporous structure by desilication and partial dealumination; note that, at a temperature of 200 °C, the reaction exhibits >99% selectivity for branched isomers and the formation of C1−C5 byproducts practically did not proceed over any of the analyzed catalysts (Figure 8). The sole partial dealumination significantly improved the accessibility of the OH groups (∼90%) in the mordenite channels, but the benefit was to a significant extent offset by a significant decrease in their concentration. The introduction of a mesoporous structure via subsequent desilication by alkaline treatment and partial dealumination by diluted acid can yield a similar percentage of accessible OH groups to dealumination, but with slightly higher concentrations. Desilication alone yielded mesopores 5− 20 nm in size and did not lead to significant exposure of the OH groups to the reactant molecules, and subsequent partial dealumination is also a necessary step in obtaining highly accessible OH groups. This necessarily leads to a reduction in the concentration of OH groups and limits their maximum available concentration. The high accessibility of acid sites in micromesoporous mordenite obtained by sequential alkaline and mild acid treatments is connected with catalytic performance exceeding that of microporous mordenite; however, the improvements in the yields are slight. To attain more significant improvement in the activities, a mordenite structure characterized by a higher concentration of Brønsted sites readily accessible for C6 molecules is crucial. A micromesoporous structure characterized by small mesopores and a high concentration of readily accessible acid sites was obtained by desilication of fluorinated mordenite zeolite. This treatment results in markedly improved catalyst activity and a large increase in the yield of branched isomers (Figures 8 and 10). Thus, it is clear that the micromesoporous structure with a high density of readily accessible OH groups is ultimately connected with catalytic properties greatly exceeding those of microporous or partially dealuminated mordenite zeolite. The concentrations of accessible Brønsted sites were correlated with the activity measured for all of the micromesoporous Pt/H-MOR zeolites (Figure 11). The yields of branched C6 isomers clearly increase with the concentrations of the Brønsted sites accessible for nhexane molecules. We also analyzed the relationships between

Figure 11. Yields of branched C6 isomers in the hydroisomerization of n-hexane for Pt/H-MOR zeolites with different concentrations of the accessible Brønsted sites. 5793

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis the concentrations of the Brønsted sites in 8-rings or 12-rings estimated using analysis of the FTIR absorption bands of the bridging OH groups in zeolites (Figure S5 in the Supporting Information) and the hydroisomerization activity (Figure S10 in the Supporting Information). The plots of the yields of branched C6 isomers against the concentrations of the Brønsted sites in 8-ring or 12-ring channels did not show any direct relationship. According to Chiang et al.,8 the higher activity of the protonic sites in the 8-ring pockets of H-MOR is associated with a specific deformation and partial confinement of the C6 molecule in the 8-ring. In the case of MOR zeolites with hierarchical porosity, the local structure and mainly the openings of the 8-ring pockets are largely affected (see Figure 6) and the specific activity of corresponding protonic sites could be altered. Summing up, the analysis of the hydroisomerization over micromesoporous Pt/H-MOR zeolites exhibiting quite different levels of hierarchical structure and the overall concentrations of acid sites, and thus the different concentrations of accessible Brønsted sites, shows that the hierarchical porosity has a promotional effect on the activity of acid sites even for the relatively small C6 molecules. The increase in the yields of branched isomers over the micromesoporous structure with a high concentration of readily accessible Brønsted sites enables a shift of the operation window to lower temperatures with more favorable thermodynamic equilibrium for branched isomers.



AUTHOR INFORMATION

Corresponding Author

*E-mail for P.S.: [email protected]. ORCID

Petr Sazama: 0000-0001-7795-2681 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the Grant Agency of the Czech Republic under Project No. GA15-12113S. The authors acknowledge the assistance provided by the Research Infrastructures NanoEnviCz (Project No. LM2015073) and ProNanoEnviCz (Project No. CZ.02.1.01/0.0/0.0/16_013/ 0001821), supported by the Ministry of Education, Youth and Sports of the Czech Republic.



CONCLUSIONS The critical function of the micromesoporous structure and the concentration and accessibility of Brønsted sites for hydroisomerization of linear hexane was elucidated by using three series of micromesoporous mordenite zeolites differing in their mesoporosity and concentration and the accessibility of acid sites. The hydroisomerization of n-hexane over microporous Pt/H-mordenite is limited by mass transfer effects due to the low accessibility of acid sites in 8-ring channels and the diffusion-restricted pseudomonodimensional channel structure. The introduction of secondary mesoporosity into mordenite using the alkaline−acid, acid−alkaline−acid, and fluorination− alkaline postsynthesis treatments proposed in this study effectively shortens the length of the 12-ring main channels, opens the side 8-ring channel pockets, provides a high concentration of Brønsted sites, and enhances the accessibility of the acid sites for hexane molecules. The shortening of the main 12-ring channels and a larger number of channel openings provide an increase in the selectivity, limiting the nonselective subsequent cracking reactions and consequently increasing the yield of the desired products at higher temperatures. The micromesoporous mordenite structure with the Brønsted sites in the environment of the micropores and accessible through mesopores alleviates the diffusion limitations due to restricted access and slow transport to/from the active site, resulting in a significant increase in the catalyst efficiency and improvements in the selectivity in skeletal hydroisomerization of n-hexane into branched isomers. It enables efficient isomerization in a temperature region that is thermodynamically more favorable for desired dibranched isomers.



Description of the structure of parent mordenite zeolite, SEM images, X-ray diffraction patterns, nitrogen adsorption and desorption on a log P scale, and 27Al MAS NMR spectra for the micromesoporous mordenite zeolites, illustration of analysis of the FTIR absorption bands of the bridging OH groups of dehydrated zeolites and the CN groups of adsorbed d3-acetonitrile, illustration of analysis of the OH group accessibility using FTIR spectra of adsorbed n-hexane, TEM images and FTIR spectra of CO adsorbed on the reduced microporous and micromesoporous Pt/H-MOR, and supplementary kinetic data (PDF)



REFERENCES

(1) Iglesia, E.; Soled, S. L.; Kramer, G. M. J. Catal. 1993, 144, 238− 253. (2) Iglesia, E.; Barton, D. G.; Soled, S. L.; Miseo, S.; Baumgartner, J. E.; Gates, W. E.; Fuentes, G. A.; Meitzner, G. D. Selective isomerization of alkanes on supported tungsten oxide acids. Stud. Surf. Sci. Catal. 1996, 101, 533−542. (3) Chica, A.; Corma, A. J. Catal. 1999, 187, 167−176. (4) Ono, Y. Catal. Today 2003, 81, 3−16. (5) Tromp, M.; van Bokhoven, J. A.; Garriga Oostenbrink, M. T.; Bitter, J. H.; de Jong, K. P.; Koningsberger, D. C. J. Catal. 2000, 190, 209−214. (6) Gounder, R.; Iglesia, E. Angew. Chem., Int. Ed. 2010, 49, 808−811. (7) Bhan, A.; Iglesia, E. Acc. Chem. Res. 2008, 41, 559−567. (8) Chiang, H.; Bhan, A. J. Catal. 2011, 283, 98−107. (9) Van Laak, A. N. C.; Sagala, S. L.; Zecevic, J.; Friedrich, H.; De Jongh, P. E.; De Jong, K. P. J. Catal. 2010, 276, 170−180. (10) Centi, G.; Perathoner, S. Coord. Chem. Rev. 2011, 255, 1480− 1498. (11) Groen, J. C.; Zhu, W.; Brouwer, S.; Huynink, S. J.; Kapteijn, F.; Moulijn, J. A.; Pérez-Ramírez, J. J. Am. Chem. Soc. 2007, 129, 355−360. (12) Gobin, O. C.; Reitmeier, S. J.; Jentys, A.; Lercher, J. A. J. Phys. Chem. C 2011, 115, 1171−1179. (13) Park, D. H.; Kim, S. S.; Wang, H.; Pinnavaia, T. J.; Papapetrou, M. C.; Lappas, A. A.; Triantafyllidis, K. S. Angew. Chem., Int. Ed. 2009, 48, 7645−7648. (14) Ferreira, A. F. P.; Mittelmeijer-Hazeleger, M. C.; Granato, M. A.; Martins, V. F. D.; Rodrigues, A. E.; Rothenberg, G. Phys. Chem. Chem. Phys. 2013, 15, 8795−8804. (15) Fan, W.; Snyder, M. A.; Kumar, S.; Lee, P. S.; Yoo, W. C.; McCormick, A. V.; Lee Penn, R.; Stein, A.; Tsapatsis, M. Nat. Mater. 2008, 7, 984−991. (16) Hartmann, M. Angew. Chem., Int. Ed. 2004, 43, 5880−5882. (17) Inayat, A.; Knoke, I.; Spiecker, E.; Schwieger, W. Angew. Chem., Int. Ed. 2012, 51, 1962−1965.

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acscatal.7b01696. 5794

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795

Research Article

ACS Catalysis (18) Perez-Ramirez, J. Nat. Chem. 2012, 4, 250−251. (19) Sazama, P.; Sobalik, Z.; Dedecek, J.; Jakubec, I.; Parvulescu, V.; Bastl, Z.; Rathousky, J.; Jirglova, H. Angew. Chem., Int. Ed. 2013, 52, 2038−2041. (20) Sun, J. L.; Bonneau, C.; Cantin, A.; Corma, A.; Diaz-Cabanas, M. J.; Moliner, M.; Zhang, D. L.; Li, M. R.; Zou, X. D. Nature 2009, 458, 1154−1157. (21) Choi, M.; Na, K.; Kim, J.; Sakamoto, Y.; Terasaki, O.; Ryoo, R. Nature 2009, 461, 246−249. (22) Zhang, X. Y.; Liu, D. X.; Xu, D. D.; Asahina, S.; Cychosz, K. A.; Agrawal, K. V.; Al Wahedi, Y.; Bhan, A.; Al Hashimi, S.; Terasaki, O.; Thommes, M.; Tsapatsis, M. Science 2012, 336, 1684−1687. (23) Jacobsen, C. J. H.; Madsen, C.; Houzvicka, J.; Schmidt, I.; Carlsson, A. J. Am. Chem. Soc. 2000, 122, 7116−7117. (24) De Jong, K. P.; Zecevic, J.; Friedrich, H.; De Jongh, P. E.; Bulut, M.; Van Donk, S.; Kenmogne, R.; Finiels, A.; Hulea, V.; Fajula, F. Angew. Chem., Int. Ed. 2010, 49, 10074−10078. (25) Groen, J. C.; Moulijn, J. A.; Perez-Ramirez, J. J. Mater. Chem. 2006, 16, 2121−2131. (26) Góra-Marek, K.; Tarach, K.; Tekla, J.; Olejniczak, Z.; Kuśtrowski, P.; Liu, L.; Martinez-Triguero, J.; Rey, F. J. Phys. Chem. C 2014, 118, 28043−28054. (27) Paixao, V.; Carvalho, A. P.; Rocha, J.; Fernandes, A.; Martins, A. Microporous Mesoporous Mater. 2010, 131, 350−357. (28) Wichterlova, B.; Tvaruzkova, Z.; Sobalik, Z.; Sarv, P. Microporous Mesoporous Mater. 1998, 24, 223−233. (29) Fyfe, C. A.; Feng, Y.; Grondey, H.; Kokotailo, G. T.; Gies, H. Chem. Rev. (Washington, DC, U. S.) 1991, 91, 1525−1543. (30) Engelhardt, G.; Lohse, U.; Lippmaa, E.; Tarmak, M.; Magi, M. Z. Anorg. Allg. Chem. 1981, 482, 49−64. (31) Eder, F.; Stockenhuber, M.; Lercher, J. A. J. Phys. Chem. B 1997, 101, 5414−5419. (32) Zholobenko, V. L.; Makarova, M. A.; Dwyer, J. J. Phys. Chem. 1993, 97, 5962−5964. (33) Bordiga, S.; Lamberti, C.; Geobaldo, F.; Zecchina, A.; Palomino, G. T.; Arean, C. O. Langmuir 1995, 11, 527−533. (34) Wakabayashi, F.; Kondo, J.; Wada, A.; Domen, K.; Hirose, C. J. Phys. Chem. 1993, 97, 10761−10768. (35) Alberti, A. Zeolites 1997, 19, 411−415. (36) Eder, F.; Stockenhuber, M.; Lercher, J. A. J. Phys. Chem. B 1997, 101, 5414−5419. (37) Viswanadham, N.; Kumar, M. Microporous Mesoporous Mater. 2006, 92, 31−37. (38) Lietz, G.; Schnabel, K. H.; Peuker, C.; Gross, T.; Storek, W.; Volter, J. J. Catal. 1994, 148, 562−568. (39) Remy, M. J.; Stanica, D.; Poncelet, G.; Feijen, E. J. P.; Grobet, P. J.; Martens, J. A.; Jacobs, P. A. J. Phys. Chem. 1996, 100, 12440−12447. (40) Chen, T. H.; Wouters, B.; Grobet, P. Chem. J. Internet 2000, 2, 7−13. (41) Chen, T. H.; Wouters, B. H.; Grobet, P. J. Eur. J. Inorg. Chem. 2000, 2000, 281−285. (42) Wouters, B. H.; Chen, T.; Grobet, P. J. J. Phys. Chem. B 2001, 105, 1135−1139. (43) Chen, J.; Chen, T.; Guan, N.; Wang, J. Catal. Today 2004, 93− 95, 627−630. (44) Samoson, A.; Lippmaa, E.; Engelhardt, G.; Lohse, U.; Jerschkewitz, H. G. Chem. Phys. Lett. 1987, 134, 589−592. (45) Sazama, P.; Wichterlova, B.; Dedecek, J.; Tvaruzkova, Z.; Musilova, Z.; Palumbo, L.; Sklenak, S.; Gonsiorova, O. Microporous Mesoporous Mater. 2011, 143, 87−96. (46) Sayed, M. B.; Kydd, R. A.; Cooney, R. P. J. Catal. 1984, 88, 137−149. (47) Kustov, L. M.; Kazansky, V. B.; Beran, S.; Kubelkova, L.; Jiru, P. J. Phys. Chem. 1987, 91, 5247−5251. (48) Fritz, P. O.; Lunsford, J. H. J. Catal. 1989, 118, 85−98. (49) Groen, J. C.; Jansen, J. C.; Moulijn, J. A.; Pérez-Ramírez, J. J. Phys. Chem. B 2004, 108, 13062−13065.

(50) Svelle, S.; Sommer, L.; Barbera, K.; Vennestrøm, P. N. R.; Olsbye, U.; Lillerud, K. P.; Bordiga, S.; Pan, Y. H.; Beato, P. Catal. Today 2011, 168, 38−47. (51) Giudici, R.; Kouwenhoven, H. W.; Prins, R. Appl. Catal., A 2000, 203, 101−110. (52) Van Geem, P. C.; Scholle, K. F. M. G. J.; Van Der Velden, G. P. M.; Veeman, W. S. J. Phys. Chem. 1988, 92, 1585−1589. (53) Muller, M.; Harvey, G.; Prins, R. Microporous Mesoporous Mater. 2000, 34, 135−147. (54) Silaghi, M.-C.; Chizallet, C.; Sauer, J.; Raybaud, P. J. Catal. 2016, 339, 242−255. (55) Yu, L.; Huang, S.; Miao, S.; Chen, F.; Zhang, S.; Liu, Z.; Xie, S.; Xu, L. Chem. - Eur. J. 2015, 21, 1048−1054. (56) Qin, Z.; Lakiss, L.; Gilson, J. P.; Thomas, K.; Goupil, J. M.; Fernandez, C.; Valtchev, V. Chem. Mater. 2013, 25, 2759−2766. (57) Pérez-Ramírez, J.; Christensen, C. H.; Egeblad, K.; Christensen, C. H.; Groen, J. C. Chem. Soc. Rev. 2008, 37, 2530−2542. (58) Milina, M.; Mitchell, S.; Crivelli, P.; Cooke, D.; Pérez-Ramírez, J. Nat. Commun. 2014, 5, 3922. (59) Reitmeier, S. J.; Gobin, O. C.; Jentys, A.; Lercher, J. A. Angew. Chem., Int. Ed. 2009, 48, 533−538. (60) Ribeiro, F.; Marcilly, C.; Guisnet, M. J. Catal. 1982, 78, 267− 274. (61) Van De Runstraat, A.; Kamp, J. A.; Stobbelaar, P. J.; Van Grondelle, J.; Krijnen, S.; Van Santen, R. A. J. Catal. 1997, 171, 77−84. (62) Stakheev, A. Y.; Shpiro, E. S.; Tkachenko, O. P.; Jaeger, N. I.; Schulz-Ekloff, G. J. Catal. 1997, 169, 382−388. (63) Kubanek, P.; Schmidt, H. W.; Spliethoff, B.; Schüth, F. Microporous Mesoporous Mater. 2005, 77, 89−96. (64) Monteiro, R.; Ania, C. O.; Rocha, J.; Carvalho, A. P.; Martins, A. Appl. Catal., A 2014, 476, 148−157. (65) Zecevic, J.; Vanbutsele, G.; De Jong, K. P.; Martens, J. A. Nature 2015, 528, 245−254. (66) Sazama, P.; Kaucky, D.; Moravkova, J.; Pilar, R.; Klein, P.; Pastvova, J.; Tabor, E.; Sklenak, S.; Jakubec, I.; Mokrzycki, L. Appl. Catal., A 2017, 533, 28−37. (67) Monteiro, R.; Ania, C. O.; Rocha, J.; Carvalho, A. P.; Martins, A. Appl. Catal., A 2014, 476, 148−157. (68) Leng, K.; Wang, Y.; Hou, C.; Lancelot, C.; Lamonier, C.; Rives, A.; Sun, Y. J. Catal. 2013, 306, 100−108. (69) Nesterenko, N. S.; Thibault-Starzyk, F.; Montouillout, V.; Yuschenko, V. V.; Fernandez, C.; Gilson, J. P.; Fajula, F.; Ivanova, I. I. Microporous Mesoporous Mater. 2004, 71, 157−166. (70) Haag, W. O.; Lago, R. M.; Weisz, P. B. Faraday Discuss. Chem. Soc. 1981, 72, 317−330.

5795

DOI: 10.1021/acscatal.7b01696 ACS Catal. 2017, 7, 5781−5795